The amyloid hypothesis states that altered Abeta production or clearance leading to gradual accumulation of aggregated Abeta, which in turn initiates the cascade of events leading to Alzheimer's disease. Increasing evidence supports the concept that the highly amyloidogenic Abeta42 isoform is the pathogenic species. Consequently, selective targeting of Abeta42 may be an excellent anti-amyloid strategy for Alzheimer's therapeutics. In the current funding cycle, we have reported that a subset of nonsteroidal anti-inflammatory drugs (NSAIDs) selectively lower Abeta42 while increasing shorter Abeta species without altering overall Abeta levels, an activity which we term gamma-secretase modulating action. Further, these findings led us io hypothesize that this activity may in part explain the apparent AD risk reduction in chronic users of NSAIDs. Thus, NSAIDs represent the prototypic members of a class of gamma-secretase modulators [GSM) that selectively lower Abeta42 in vitro and in vivo. We also provided preliminary characterization of its potential mechanism of action, which we now believe is centered on the gamma-secretase complex itself, including the APR substrate. In the next funding cycle, we hypothesize that this gamma-secretase modulating activity may target the initial epsilon-cleavage site, which appears to precede and possibly predict the cleavage of various Abeta peptides at the gamma-site. We will test this hypothesis in cultured cells and with in vitro membrane preparations. Further, in studies in collaboration with Project 2, we will examine the biology of shorter Abeta peptides that are invariably increased by Abeta42 lowering agents. We will ask whether shorter Abeta peptides are neurotoxic and whether they modulate the toxicity and aggregability of Abeta42 peptide by testing synaptic transmission and synaptic plasticity following exposure to these peptides. Finally, we will test whether combination treatments with Abeta42 lowering GSMs and a second unrelated compound are superior to treatment with single agents in reducing amyloid levels and amyloid associated pathology in APP transgenic mice. We hypothesize that targeting multiple cellular pathways will ultimately be more efficacious than single targets. We hope that our studies will be translated to testing in AD subjects in the future if we can demonstrate synergy with combination treatments.